Identifying the Information Contained in a Flawed Theory

Abstract

One common approach to using a prior domain theory as a learning bias is to revise the theory in accordance with a set of training examples. More recently, another class of methods has arisen in which the theory is reinterpreted, either by probabilizing it, or by using its components in constructive induction. Revision-based methods tend to work best when flaws in the given theory are localized, whereas reinterpretation methods tend to work well when flaws are distributed evenly throughout the theory. This paper describes a `meta-learning' algorithm which, given a flawed domain theory, determines the general nature of the theory's flaws by analyzing the information flow in the theory. The method works by first `probabilizing' the theory, and then selectively `de-probabilizing' components, based on the theory's performance on a preclassified set of training examples. This method distinguishes between those parts of the theory which should be interpreted as given and those which need to ...